sign in sign up
<<
Home , Telecommunication

International Journal of 7 (2013), 2285–2311 1932–8036/20130005

Determinants of Mobile Adoption in Kurdistan

NABAZ T. KHAYYAT1 Seoul National University

ALMAS HESHMATI Sogang University

This study analyzes Kurdistan’s potential for effectively using mobile telecommunication. It identifies key factors determining the adoption of mobile telecommunication . A conceptual model is specified, and several hypotheses are tested with a sample of 1,458 Kurdish users in 2010. A discrete choice methodology is used to test two models of mobile telecommunication acceptance: choice of service providers and usage pattern. The results indicate that Korek is the favorite service provider in the Kurdistan region of Iraq, and the subscribers mostly use the service for work purposes. The findings have implications for competition in the market and flows of investment resources to targeted market segments with potential expansion.

Keywords: multinomial logit model, mobile telecommunication, adoption, service providers, Kurdistan, Iraq

Introduction

The Kurdistan Region is an autonomous region in federal Iraq that was cut off from all basic postal and services after the first Gulf War in 1991. The region became dependent on - based and communication systems and services. The expansion and diffusion of these services have been rapid and promoted by the Kurdistan Regional Government (Khayyat, 2010; Khayyat & Heshmati, 2012).

Cell phone and connectivity and use have increased dramatically in recent years. In a short time, information and communication technology (ICT) has become one of the main interests of the Kurdistan government and business community. They have made significant inroads in both public and private sectors of Kurdish society by incorporating the new into governance and business activities. Despite significant progress and public investment in the , the Kurdistan region is

1 The authors wish to thank two anonymous referees and an editor of the journal for their very useful comments and suggestions on an earlier version of this manuscript and Soran University for financial support.

Nabaz T. Khayyat: [email protected] Almas Heshmati: [email protected] Date submitted: 2012–01–31

Copyright © 2013 (Nabaz T. Khayyat & Almas Heshmati). Licensed under the Creative Commons Attribution Non-commercial No Derivatives (by-nc-nd). Available at http://ijoc.org. 2286 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

a latecomer in the area of ICT because of political issues and economic, social, and cultural causes such as language barriers and lack of training and development programs.

With a population of nearly 5.2 million—36% aged 0 to 14 years, 60% aged 15 to 64, and 4% over age 64—the diffusion rate in the telecom sector is between 75% and 78% (UNDP, 2004). The diffusion rate is lower than that of neighboring countries; for example. Kuwait and Saudi Arabia’s mobile penetration rate reached 150% by the end of 2012 (ITU, 2012). The three mobile phone operators with GSM licenses in Iraq and Kurdistan—Zain Iraq, AsiaCell, and Korek—are relatively new, and their rapid growth nationwide needs to be tracked and studied to derive the necessary policy implications for their business. The telecom sector in Iraq and Kurdistan has witnessed an average of 11.6% growth in the subscription rate, as the rate of subscription increased from 5% in 2005 to 75% in 2012. Zain Iraq has the largest share in the market with 53%, followed by AsiaCell with 31% and Korek with 16% (Marcopolis, 2012).

The objective of this research is to study the factors that influence the adoption of mobile telecommunication service in Iraqi Kurdistan. Such measures are important for the telecommunication sector to serve customers’ needs and ensure business sustainability and economic growth. The direct contribution is to identify and estimate the factors that affect the adoption of mobile telecommunication service in the region—a developing area with little infrastructure for wired telephone and basic postal services after the first Gulf War in 1991. The results are useful to other regions with similar conditions. In addition to its insights for the management and regulation of the telecommunication sector, this study has significant potential for education and research in the region.

Literature Review

Mobile telecommunication services have changed people’s lifestyles by transforming and revolutionizing the way they work and interact. In addition, they have provided opportunities for a powerful communication medium to influence business and social life.

Many studies have examined the adoption of mobile services and systems, and various models have been applied to predict customers’ intentions to use technological innovations (Alborz, 2009; AlHinai, Kurnia, & Johnston, 2007; Baker & Bellordre, 2003; Bouwman & van de Wijngaert, 2009; Carlsson, Hyvonen, Repo, & Walden, 2005; Coursaris, Hassanein, Head, & Bontis, 2007; Hong & Tam, 2006; Hsu, Lu, & Hsu, 2007; H. W. Kim, Chan, & Gupta, 2007; I. Lee, Choi, Kim, & Hong, 2007; Lu, Yao, & Yu, 2005; Park, Roman, Lee, & Chung, 2009; Sanford & Oh, 2010; Yang & Jolly, 2009). We rely on these theoretical models to determine the relevant factors that predict the intention to adopt a mobile phone. Among the most significant and applicable theories to study intentions to adopt mobile phone technology is the technology acceptance model (TAM) developed by Davis (1989). Other theories include the theory of reasoned action developed by Fishbein, I., and Belief (1975), and the theory of planned behavior by Ajzen (1991). Several studies rely on these three theories’ models in explaining the intention to adopt ICT systems (Venkatesh, Morris, Davis, & Davis, 2003).

International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2287

The Technology Acceptance Model

The TAM was developed to explain potential users’ behavioral intentions in using technological innovation. It is based on the theory of reasoned action (Fishbein & Ajzen, 1975), in which attitudes that are influenced by beliefs form behavioral intentions. The TAM uses this causal sequence of belief–attitude– intention–behavior to understand the determinants of acceptance, which then triggers actual behaviors (W. J. Lee, Kim, & Chung, 2002). The TAM describes the causal relationships between system design features, perceived usefulness, perceived ease of use, attitude toward using, and actual usage behavior. The TAM provides an informative representation of the mechanisms by which design choices influence user acceptance.

Two key constructs are perceived usefulness and perceived ease of use. Perceived usefulness refers to the prospective user’s subjective notions about how much using a specific application system will increase his or her job performance in an organizational context, which relates to extrinsic characteristics such as efficiency and effectiveness. Perceived ease of use refers to the degree to which the prospective user expects the target system to be of effort, which relates to intrinsic characteristics such as ease of use and flexibility. These two perceptions combine to affect the adoption of new technologies (Adams, Nelson, & Todd, 1992).

External factors such as individual favorites or constraints resulting from certain situations also are expected to affect users’ acceptance of the technology when these factors are within the two key construct of perceived usefulness and perceived ease of use. According to King and He (2006), perceived usefulness might affect actual use irrespective of attitude if the use of the technology offers direct benefit to the user.

Despite several studies confirming the robustness of the acceptance models, many scholars have identified some important limitations associated with using such models. Chuttur (2009) classified the criticism of the acceptance models into three categories: First, the methodology relies on self-reported use data instead of actual use data. This subjective measure is considered by some scholars unreliable for measuring the actual use of a system. Second, the variables and the relationships constructed in the acceptance models are limited in terms of their ability to be replicated. Many scholars have found that perceived usefulness has more influence on system acceptance than perceived ease of use. Therefore, these two sets of variables may not mediate all relevant influences from external environmental factors on the system usage. As a result, some external factors such as level of education and age also may have direct influences on system usage (Burton-Jones & Hubona, 2006). Finally, according to Bagozzi (2008), behavior cannot be considered as a terminal goal. He has argued that the acceptance models are deterministic and not suitable for explaining and predicting system usage. Bagozzi notes the need for a paradigm shift because of critical shortcomings in the paradigm. Bagozzi’s proposition was that there might be other factors that alter the dominant paradigm.

Based on these theories, researchers have studied many aspects of mobile phone adoption, such as the effects of factors such as usefulness, ease of use, enjoyment of using a service, content and system quality, and impact of technical issues. Despite possible individual limitations but strong complementarities, the experience gained from the studies and their results can be used to specify suitable models of adoption of mobile telecommunication service in Kurdistan. 2288 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

Technology Adoption in Developing Countries

Fong (2009) proposed some “relevant” factors for ICT adoption in developing countries such as market structure, institutional capacity, socioeconomic factors, human capabilities, government policies, basic utility infrastructure, stakeholders’ interactions in the system, and international cooperation and collaboration.

An empirical examination of the factors affecting consumers’ intentions to use mobile commerce in proposed an extension to TAM. Five factors were examined: perceived usefulness, perceived ease of use, social influence, perceived cost, and trust (Wei, Marthandan, Chong, Ooi, & Arumugam, 2009).

In research on technological adoption in poor countries, Foster and Rosenzweig (2010) recommend further study focusing on the interaction between behavior, market settings, traditional institutions, and technology payoffs to understand the different experiences over time for different countries in adopting new technologies. Gulbahar and Guven (2008) surveyed 326 primary school teachers in Turkey to examine the use and adoption of ICT. They found that, despite the teachers’ willingness to use ICT, accessing the ICT tools was difficult.

Other studies examining the ICT adoption in developing countries have added to the growing stock of knowledge in the field of ICT (Aboelmaged, 2010; Al-Qirim, 2007; Al-Shafi & Weerakkody, 2009; Avgerou, 2009; Bwalya, 2009; Dercon & Christiaensen, 2011; Duncombe & Boateng, 2009; Giné & Yang, 2009; Howard & Mazaheri, 2009; Khayyat & Heshmati, 2012; Kijsanayotin, Pannarunothai, & Speedie, 2009; Park et al., 2009). A strand of literature has changed focus from theories explaining adoption intention to theories of usage of technology after adoption (Bhattacherjee & Premkumar, 2004; S. S. Kim & Malhotra, 2005).

Conceptual Framework and Research Hypothesis

The model presented here includes five main determinants: perceived usefulness, perceived ease of use, perceived enjoyment, price, and demographic characteristics such as age, gender, location, education level, and occupation. Type of service provider and purpose of using a mobile phone are added as dependent variables against which all other variables are analyzed.

Perceived usefulness is defined by Venkatesh and Davis (2000) as “the extent to which a person believes that using the system will enhance his or her work performance” (p. 187). Previous research found that perceived usefulness is the strongest predictor of intention to use and purchase the technology. We expect such an influence will persist under the current circumstance; therefore, our first hypothesis is:

H1: Perceived usefulness of affects an individual’s intention to adopt the technology.

Perceived ease of use is defined as the degree to which interaction with the technology is clear and understandable (Ndubisi, Gupta, & Massoud, 2003). Perceived ease of use, as one of the characteristics of TAM, has been found to be an important factor influencing the intention to purchase a technology (Davis, 1989). Thus, our second hypothesis is: International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2289

H2: Perceived ease of use of the mobile technology affects an individual’s intention to adopt the technology.

Perceived enjoyment is defined by Davis, Bagozzi, and Warshaw (1992) as “the extent to which the activity of using the technology is perceived to be enjoyable in its own right, apart from any performance consequences that may be anticipated” (p. 1113). The concept is also used by others who argue that it affects the intension to use the technology (Moon & Kim, 2001; van der Heijden, 2003; Venkatesh & Davis, 2000). Accordingly, we propose the following hypothesis:

H3: Perceived enjoyment of the mobile technology affects an individual’s intention to adopt the technology.

Rogers (1995) has presumed intrinsic generalizations concerning characteristics of the adopter that influence innovativeness: socioeconomic status, personality values, education level, and communication behavior. The age characteristic, according to Rogers’ generalization, do not distinguish the earlier adopters from the later adopters. However, some studies have shown that early adopters are younger, and some have shown that they are older.

We propose the following additional hypotheses regarding the adoption and usage of mobile telecommunication service. Because these variables can be considered control variables, no theoretical justifications for their use are needed.

H4: An individual’s age influences his or her intentions to adopt the technology.

H5: An individual’s gender influences his or her intentions to adopt the technology.

H6: An individual’s occupation influences his or her intentions to adopt the technology.

H7: An individual’s educational level influences his or her intentions to adopt the technology.

H8: An individual’s location influences his or her intentions to adopt the technology.

H9: An individual’s income influences his or her intentions to adopt the technology.

Studies on the telecommunication business show that customers are sensitive to the service price. Price is used to attract new and current customers from competitors through marketing campaigns (Roos & Gustafsson, 2007). The final hypothesis, then, is stated as follows:

H10: Adoption of is related to the price of the purchased service.

2290 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

In Kurdistan, the service price is set by the service provider according to the market competition. There are no regulations about price setting in the region.

The five main determinants of mobile phone technology adoption used in this study are presented in Figure 1. The figure also provides a graphic representation and summary of this study’s conceptual foundation.

Perceived usefulness H1

Perceived ease H2 of use

H3 Perceived Intention to enjoyment H4, H5, H6, H7, H8, H9 purchase

Demography

H10

Price

Figure 1. A conceptual model for consumer acceptance of mobile phone technology.

International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2291

Data, Models Estimation, and Results

Mobile Service Survey

To study the mobile telecommunication service network subscribers in Kurdistan, the considered sample was limited to the general public living in the region. To collect a representative sample of respondents,2 different methods were applied for data collection. The two main mobile phone operators in the region (Korek and AsiaCell) collected each 400 surveys in 2010 by calling their customers randomly and conducting the questionnaire directly through phone calls. The surveys were conducted upon request from the principal author. In addition, 658 forms were collected by the Ministry of Transportation and Communication and the Ministry of Agriculture and Water Resources targeting government employees. The survey distribution reflected the population structure in the region’s governorates.

A total of 1,458 questionnaires were received. The median respondent was between 26 and 40 years of age, hold a diploma or bachelor degree, and be a public-sector employee with an annual disposable income of less than US$8,000. The distribution and descriptive statistics of the main variables are reported in Table 1.

Table 1. Frequency (Main Variables) of the Survey Data (N = 1,458).

Variable No. of Obs. % Variable No. of Obs. % Location Use purpose

1 555 38 1 824 56

2 236 16 2 465 31

3 667 45 3 80 5 Gender 4 89 6

1 780 53 Hours used

2 678 46 1 963 66 Age 2 380 26

1 486 33 3 115 7 2 578 39 phone brand

3 344 23 1 1,007 69

4 50 3 2 200 13 Education 3 83 5

1 174 11 4 50 3

2 The sample is representative according to the 2007 statistical yearbook released by KRG Ministry of Planning. See http://www.krso.net/documents/99/FactSheet1_Demography_21072012.pdf

2292 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

2 163 11 5 55 3

3 355 24 6 63 4 4 685 46 Preferred features

5 81 5 1 215 14 Occupation 2 205 14

1 70 4 3 173 11

2 160 10 4 499 34

3 800 54 5 366 25 4 428 29 User satisfaction

Income 1 972 66

1 165 11 2 486 33 2 632 43 Service quality

3 345 23 1 406 27

4 98 6 2 548 37

5 68 4 3 333 22

6 64 4 4 124 8

7 86 5 5 47 3 Operator 1 654 44

2 748 51

3 56 3

The questionnaire consisted of two parts: Part 1 contained questions about consumer characteristics and demography, and part 2 contained questions about the driving factors of the customer’s technology adoption. The questions were designed to discover customers’ perceptions about customer service after experiencing it and to inquire about the brand and type of cell phone purchased after the technology adoption. Based on previous research (Liao, Tsou, & Huang, 2007; Moore & Benbasat, 1991; Pagani, 2004), perceived usefulness is measured with questions about the daily amount of time the service is used, the main purpose of use, and the network quality. Perceived ease of use is measured with questions about the customers’ preferred cell phone features; the questions were adapted from Agarwal and Prasad (1997).

The questionnaire was designed in English and Kurdish versions. The English version was created first, because this was easier and more convenient to adapt from the original format initiated by Parasuraman, Zeithaml, and Berry (1985). The English questionnaire was then translated into the Kurdish language. Table 2 presents definitions of the variables and their scales.

International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2293

Table 2. Descriptions of the Main Variables and Their Classifications.

Main Subvariable Description Type of Construct variable variable Location Erbil Consumer residing in the capital of the Erbil region Dummy Dhok Consumer residing in the of Duhok (1 for yes, Suli Consumer residing in the city of 0 otherwise) Sulaymaniyah Gender Male Male consumer Female Female consumer Age Age Age of consumer Continuous Education Prims Level of education up to primary school Mids Level of education up to middle school Hs Level of education up to high school

College Level of education: bachelor or diploma

Postgrad Level of education: high diploma, master, Demography doctorate

College Level of education: bachelor or diploma Dummy Occupation None Occupation: none/retired (1 for yes, Private Occupation: self-employed 0 otherwise) Public Occupation: public employee Student Occupation: student Income Income Monthly income Continuous Operator 1: Users of AsiaCell Discrete 2: Users of Korek variable 3: Users of other services (1, 2, 3) Name of Korek Consumers use Korek service Service service Asia Consumers use AsiaCell service Dummy resources operator Othersp Consumers use the other services (1 for yes, used (Ariafon/Mobitel) 0 otherwise) Price AsiaCell: 1.16 Korek 1.245 Cost Others on average: 1.92 Continuous Cost Cost Time use  price Purpose of Work Primary purpose of using the service: Work- Efficiency use related Social Social interaction Dummy Entmnt Entertainment (1 for yes, 0 otherwise) Otheprps A combination of purposes

2294 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

Hours used Houruse1 Amount of time service is used (receiving/making a call): less than 30 minutes per day Houruse2 Amount of time service is used (receiving/making a call): 30–60 minutes per day Houruse3 Amount of time service is used Dummy Utility (receiving/making a call): (1 for yes, More than 60 minutes per day 0 otherwise) Cell phone Name of cell phone brand: Nokia brand Korean LG or Samsung Othebnd /Motorola/other Preferred Color Color features Multimedia Battery Battery life Tech High-tech options Easyuse Ease of use User satisfaction Is the user satisfied with the current service Service quality User’s rating of service quality 1 = very high Discrete 2 = high variable 3 = medium (1, 2, 3, 4, 4 = low 5) 5 = very low

The questions were categorized according to the five main constructs: demography, service resources, cost, efficiency, and utility (see Table 2). In addition to demographic constructs, the study includes the constructs of perceived usefulness and efficacy, which include utilization rate factors. It also includes a construct of efficiency expressed in terms of the customer’s main purpose of using the mobile service. A correlation of the variables is presented in Table 3. No significant correlation exists between the demographic variables, which allows the use of the fourth and fifth determinants as control variables.

The distribution of the mobile subscribers is shown in Table 4. The number of users for Korek, AsiaCell, and others (Ariafon and Mobitel) are 748 (51.3%), 654 (44.86%), and 56 (3.84%), respectively. We pooled the number of users for the two latter operators into one due to their relatively small number and market share. The market share of the service providers as represented by our sample is close to their actual market shares.

International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2295

The Multinomial Logit Models

The Logit model is used to model a relationship between a discrete dependent variable Y and one or more independent variables X. The dependent variable, Y, is a discrete variable (if it contains only two choices, then the model is called a binomial logit; although it is called multinomial logit if the dependent variable Y is multicategorical variable) that represents a choice, or category, from a set of mutually exclusive choices or categories. For example, a researcher models the choice of mobile phone purchase (from a set of mobile phone brands), the choice mode (such as Nokia, iPhone, Samsung, LG, etc.), the features of the mobile phone (color, size, display screen, etc.), or the service provider choice (SP1, SP2, SP3, etc.). The independent variables are presumed to affect the choice or category or the choice maker, and represent a priori beliefs about the causal or associative elements important in the choice or classification process. In the case of ordinal scale variables, an ordered logit or probit model can be applied to take advantage of the additional information provided by the ordinal over the nominal scale (Greene, 2008).

To explain the concept of logistic regression, we follow the definition of Train (2009) . A decision maker, labeled i, faces a choice among J alternatives (j = 1, 2,…, J). The decision maker would obtain a certain level of utility from each alternative. This utility is denoted as Uj,. The decision maker chooses the alternative that provides the greatest utility. Our behavioral model takes the following form: A decision maker chooses alternative j if and only if Uij > Uik ∀ j ≠ k. The probability that decision maker i chooses alternative j is equal to: Pij = Prob(Uij > Uik ∀ j ≠ k). If j > 2, then the model is a multinomial choice probability model. The basic function of the logistic analysis is:

1 (1) Pij  E(Yj |1/ x jm )  , 1  exp((0  mX jmi ))

where Pij is the probability that individual i chooses alternative j, Xim is a vector of observed variables relative to alternative j, and β0 and βm are vectors of coefficients to be estimated.

In this study we have specified two models using logistic regression methodology. The two models are constructed as functions of the different data characteristics and are estimated independently.

2296 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

Table 3. Correlation Matrix, 1,458 Observations.

Locat Gender Age Educat Occup Income Spname Satisfy Handtyp Preftur Houruse Usprpos Qosrate

Locat

Gender 0.009

Age 0.186 0.176

Educat 0.068 0.042 0.166 Occup 0.019 0.021 -0.368 0.226

Income 0.015 0.031 0.109 0.176 0.027

Spname 0.577 0.042 -0.194 0.021 0.048 0.054

Satisfy 0.087 0.158 0.043 0.146 0.020 0.012 0.100

Handtyp 0.032 0.024 -0.021 0.033 0.008 0.018 0.137 0.013

Preftur 0.039 0.015 0.113 0.027 0.128 0.044 0.026 0.075 0.149

Houruse 0.017 0.006 0.015 0.092 0.059 0.500 0.112 0.002 0.126 0.027

Usprpos 0.039 0.132 0.206 0.016 0.108 0.118 0.043 0.025 0.138 0.144 0.022

Qosrate 0.021 0.089 0.005 0.013 0.020 0.104 0.075 0.302 0.004 0.022 0.044 0.148 —

Table 4. Distribution of Mobile Telephone Service Subscribers.

Percenta Service provider Frequency ge Asia Cell 654 44.86 Korek 748 51.30 Aria/MobiTel 56 3.84 Total number of observations 1,458 100.00

To examine how well the two models fit the data, the log likelihood ratio test (LR test) is used to compare one model without restriction (the full model) with the other model with restriction (the null model). The utility maximization that is subject to restriction will not imply a larger maximum when it is compared with utility maximization that is subject to no restriction (Gujarati, 2004).

The LR test is defined as follows: λ = 2(LRu  LRr), when LRu is the log likelihood for the model

with no restriction, and LRr is the log likelihood for the model with restriction. The test statistic λ follows a 2 χ distribution with degree of freedom equal to the number of restrictions imposed by the null hypothesis.

The difference between LRu and LRr indicates whether the restrictions are correct.

International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2297

The Choice of Service Operator

Because we are interested in a sample of individuals and mobile service operators, we used the multinomial logit model to study the choices of the subscribers for the mobile telecommunication service operators in Kurdistan. This can be considered a special case of a general model of utility maximization (Greene, 2008). This model will allow us to make inferences and identify the factors influencing users’ behavior in choosing the type of services available within the region’s mobile telecommunication service network.

It is assumed in the model that the independent variables consist of a set of individual user characteristics such as gender, age, income, education, and employment. These are the same for all the choices regardless of service provider chosen. The observational setting is the individual’s choice among a set of alternatives, where it is assumed that an individual’s characteristics determine his or her choice. The service provider (Sp) has three values—0 = AsiaCell, 1 = Korek, and 2 = Other provider—and is specified as follows:

SP     Age   Cost   Income   Gender   Locat i 0 Age i Cost i Income i Gender i j ji j  1 (2)    Eduat    Occup    Usprpos  j ji j ji j ji i j  1 j  1 j  1 where Age, Income, Gender, Locat, Educat, and Occup represent the age, income, gender, location, education level, and occupation of the consumer; and Cost and Usprpos represent the cost of using the service and the main purpose of using the service, respectively. The number of categories in each group of variables may differ. Because there are multiple categories for SP, we choose a base category (AsiaCell = 0) as the reference group.

The reference group for explanatory variables is comprised of subscribers with the following characteristics: women residing in Erbil, with a primary school education, are unemployed or have no permanent job, and use the phone for work-related purposes.3 To test the fit of the model, we calculated the predictive values for the three alternatives: (608 + 529 + 8)/1,458 = 78.5%, which implies that 78.5% of the choices are correctly predicted by the model (see Table 5). The results show that AsiaCell has the highest number predicted, the model estimated correctly 79% of the actual outcome. The model predicted 78% of the actual outcome for Korek service provider.

Table 5. Number of Actual and Predicted Outcomes from Model 1, the Choice of Service Provider.

Predicted alternative service provider Total Actual alternative AsiaCell: 0 Korek: 1 Other: 2 service provider AsiaCell: 0 608 (79%) 140 0 748 Korek: 1 123 529 (78%) 2 654 Other: 2 38 10 8 (80%) 56 Total 769 679 10 1,458

3 We have chosen different reference groups in the models, but this had no effect on the results. 2298 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

The likelihood ratio (LR) in this model is 814.75 (see Table 6). With 32 degrees of freedom, it indicates that the effects in model specification are statistically highly significant at the 10% level of significance. The null hypothesis (the model with only intercept) is rejected, and the independent variables are jointly significant.

According to the figures shown in table 6, we can interpret the results as follows: Residents of Sulaymaniyah and Duhok prefer to use Korek over AsiaCell (where the coefficients of SULI and DHOK, 3.168 and 1.048, respectively, are significant at the 10% level). The same effect is seen when we compare residents of Duhok who prefer AsiaCell over other service providers (but the coefficient of DHOK, 0.611, is significant at the 1% level). This supports our hypothesis (H8) that the individual’s location influences his or her adoption of a mobile telephone service.

Table 6. Parameter Estimates for Service Provider Model (N = 1,458).

Dependent Variable: Sp

Variables Y=1 Y=2 Main Sub Korek OtherSP Constant Constant -5.984*** -4.769*** Female Reference Gender Male 0.211 -0.006 Erbil Reference Location Suli 3.168*** -0.911* Dhok 1.048*** 0.611* Age Age 0.033*** 0.015 Income Income -0.0003*** -0.0001 Prims Reference Mids -0.311 -0.318 Education Hs -0.508* -2.061*** College -0.236 -1.213*** Postgrad -0.641* -0.717 None Reference Private 1.809*** 1.281** Occupation Student 1.939*** 0.050 Public 2.411*** 0.406 Work Reference Social 0.302* 0.529 Purpose of Use Entmnt 0.262 1.258** Otheprps 0.215 0.947* Cost Cost 0.0417*** 0.062*** Likelihood Ratio: LR 814.75 (32 df) Note. Standard errors are robust. * p < .10, two-sided test. ** p < .05, two-sided test. *** p < .01, two-sided test. International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2299

Older generations are more likely to use Korek than AsiaCell (a 10% level of significance with coefficient 0.033). This result supports the hypothesis (H4) that age influences the adoption of mobile phone service. H9 is also supported; consumers with lower income prefer AsiaCell over Korek, and the results are logical because AsiaCell service is cheaper than Korek service. Occupation has an overall positive impact on the adoption of service. Students and those who work in the public and private sectors are more likely to favor Korek over AsiaCell. H6 is also supported.

The multinomial logit for men relative to women, for Korek relative to AsiaCell, and other service providers relative to AsiaCell are conditional on the other variables in the model. These are held constant at 0.211 units higher for using Korek relative to AsiaCell and 0.006 units lower for using the other service providers relative to AsiaCell, respectively. However, the prediction is not reliable, because the coefficient is not statistically significant. Hence, we are unable to test gender-related H5 with this model.

People who like to use their mobile phone mainly for entertainment purposes prefer to have service providers other than Korek and AsiaCell, at a 5% level of significance with a coefficient of 1.258. This finding supports our claim, and hence H1, H2, and H3 can be justified. On the other hand, we notice an overall negative influence of education level—in particular, customers with different education levels make different choices. For example, customers with high school or a higher level of education are more likely than customers with some primary education to use AsiaCell than other service providers. H7 is justified based on these results.

By examining the impact of cost on the adoption of service, the coefficient of cost is negative and it is statistically highly significant in both groups compared to the reference group. This supports H10.

The Usage Pattern of Mobile Phone

To be accepted, a technology must satisfy fundamental usability requirements and be perceived as useful by its intended user community (Davis, Bagozzi, & Warshaw, 1989). This model analyzes the factors affecting the purpose of using the mobile telecommunication service (USPURPOS) among mobile phone subscribers in Kurdistan. This model supports our study in determining the factors relevant to mobile telephone adoption. We applied the multinomial logit model again, specified as follows:

USPURPOS i  0  Age Age i  Cost Costi  Income Income i  Gender Gender i (3)    j SPNAME ji    j BRAND ji    j FEATUREji    j HOURUSE ji  j 1 j 1 j 1 j 1

where Spname, Feature, and Houruse represent service provider name, the preferred features available in the cell phone, and the length of time the customer uses the service on a daily basis. Uspurpos has four choices (0 = work related, 1 = social, 2 = entertainment, and 3 = other purpose). The reference group for the dependent variable is the work-related group. The reference groups for the explanatory variables are as follows: women, AsiaCell for SP, Nokia for BRAND, ease of use for Feature, and houruse1 for Houruse. 2300 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

We examine how the specified model fits to explain the variations in the data survey, and Table 7 shows the number and percentages of actual and predicted observations of the dependent variable Usprpos.

Table 7. of Actual and Predicted Outcomes-Model 2, Use Purpose

Predicted alternatives Total

Actual alternative Work: 0 Social: 1 Entertainment: 2 Others: 3

Work: 0 709 (63.0%) 110 5 0 824

Social: 1 314 139 (44.5%) 11 1 465

Entertainment: 2 37 34 9 (36.0%) 0 80

Others: 3 61 28 0 0 (0.0%) 89

Total 1,121 311 25 1 1,458

Mobile service subscribers in Kurdistan are mostly using the service for their work. The model estimated 63% of the actual outcome; the second possible use is for social life, where the model estimated 44.5% of the actual outcome correctly. To test the fit of the model, we calculated the overall predictive values for the four alternatives: (709 + 139 + 9)/1,458 = 58.8%. We find that 58.8% is correctly predicted by the model estimating the purpose of using the mobile service.

International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2301

Table 8. Parameter Estimates for Usage Pattern Model (N = 1,458).

Dependent Variable: Usprpos Variables Y=1 Y=2 Y=3 Main Sub Social Entertainment Combined Constant Constant 2.834 -10.307** 2.897 Female Reference Gender Male -.918*** -0.3081 -0.500** Age Age -0.028*** -0.109*** -0.041*** Income Income -0.96D-04 -0.006*** -0.006*** Cost Cost -0.076 0.390** -0.140 Ease of Use Reference Preferred Tech -0.523*** 0.575 0.282 Features Media 0.370* 2.288*** 0.696* Battery 0.239 1.515** 0.585 Color 0.557*** 2.187*** 0.032 Houruse1 Reference Hoursuse Houruse2 1.672 -7.716** 3.152 Houruse3 2.782 -12.537* 6.248 Nokia Reference Brand Name Korean 0.077 0.691** -0.1826 Othebnd 0.681*** 1.575*** 0.793** Asiacell Reference SP NAME Korek -0.634* 1.963** -0.857 Othersp -0.279 2.653** -0.394 Likelihood Ratio LR 327.41 (42 df) Note. Standard errors are robust. * p < .10, two-sided test. ** p < .05, two-sided test. *** p < .01, two-sided test.

The likelihood ratio in this model is 327.41 with 42 degrees of freedom (see Table 8). This indicates that the effects in the model specification are statistically highly significant. With a 1% level of significance, we can reject the null hypothesis that the coefficients of the independent variables are all simultaneously zero. The results shown in Table 8 can be interpreted as follows.

The negative and statistically significant values of male, age, and income imply that women, younger cohorts, and lower-income customers are more likely to use mobile service for performing their 2302 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

work. A possible explanation for this finding is that younger people and women generally earn less income and hence they mostly use the service for necessary tasks. In Kurdistan, most employers subsidize employees’ calling expenses. As expected from the economic theory, the cost coefficient is negative and statistically significant, indicating that use of the service for entertainment decreases as the service price increases.

The more time per day the service is used, the more likely it is used for non-entertainment purposes, a claim supported by examining the coefficients of houruse2 and houruse3. Moreover, Korek service users are more likely to use the service for work than for social purposes (coefficient –0.634). This might be due to the high price of using the service in comparison with AsiaCell. In general, customers who use the service for social-related activities prefer AsiaCell over the other service providers, but for entertainment purposes, AsiaCell is less favored. This supports H3, stating a positive relationship between entertainment and adoption of the service.

Table 9. Summary of Hypotheses Tested.

Hypotheses Proved and justified in models H1, H2 Model 1 H3 Model 1, Model 2 H4 Model 2 H5 Model 2 H6 Model 1 H7 Model 1 H8 Model 1 H9 Model 1, Model 2 H10 Model 2

H4, H9, and H10 are justified in this model as well as in the previous model, which indicates the power of the two models in determining the adoption factors of the mobile service telecommunication in Kurdistan. Subscribers who use Korean cell phone brands (Samsung and LG) and other brands are more likely than Nokia subscribers to use the phone for entertainment. This implies that Korean brands have more entertainment features than Nokia. Table 9 presents a summary of the proved and justified hypothesis by the two models constructed in this article.

Marginal Effects

The estimated coefficients from multinomial logit models do not have a direct interpretation. Parameter estimates will give us only the expected change in logit, not the probability of choosing a certain service provider in the first model and the probability of using the service for a specific purpose in the second model. The purpose of this study is to examine the direct effect of variables on the probability of choosing a service provider and to examine the direct effect of variables on the probability of using the service for a specific purpose, leading us to the marginal effects. The marginal effects are reported in Table 10. International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2303

Table 10. Marginal Effects Obtained From Estimating Technology Adoption Models (N = 1,458).

Model 1: Spnam Model 2: Usprpos

Variables Y = 1 Y = 2 Y = 1 Y = 2 Y = 3

Main Sub Korek Othersp Social Entertainment Combined

Constant Constant 1.420*** 0.043** 0.644* 0.238** 0.111

Female Reference Reference Gender Male 0.052 0.002 0.191*** 0.001 0.009

Erbil Reference

Location Suli 0.781*** 0.045*** Dhok 0.250*** 0.003

Age Age 0.008*** 0 .005*** 0.002*** 0.002***

Income Income 0.00*** 0 0 0.000*** 0.000***

Prims Reference Mids 0.073 0.004

Education Hs 0.107 0.037*** College 0.047 0.022**

Postgrad 0.151 0.009

None Reference Private 0.431*** 0.010 Occupation Student 0.473*** 0.016

Public 0.585*** 0.013

Work Reference Social 0.070* 0.008 Purpose Of Use Entmnt 0.053 0.023*

Otheprps 0.045 0.017

COST Cost 0.010*** 0.001*** 0.017 0.009** 0.006

Ease Of Use Reference

Tech -0.124*** 0.015 0.023

Preferred Media 0.053 0.044*** 0.026 Features Battery 0.031 0.029** 0.024

Color 0.106** 0.041*** 0.010

2304 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

Houruse1 Reference Hours Houruse2 0.365 0.176** 0.142 Used Houruse3 0.588 0.288** 0.287

Nokia Reference

Brand Korean 0.015 0.014** 0.011 Othebnd 0.125*** 0.027*** 0.026

Asiacell Reference Service Operator Korek -0.138* 0.046** -0.035 Used Othersp -0.072 0.057** 0.018

Note. Standard errors are robust.* p < .01, two-sided test. ** p < .05, two-sided test. *** p < .10, two-sided test

The marginal effects are interpreted as follows. The coefficient of the residents of Suli (Sulaymaniyah City) and Dhok (Duhok City) in Model 1 for Korek are 0.781 and 0.250, which implies that residents of Sulaymaniyah and Duhok are more likely to choose Korek. The mountainous topography of Sulaymaniyah and Duhok means that a service provider with strong call will have a better chance of adoption, and Korek, indeed does have better call signals than other service providers.

The probability of older generations to choose Korek over AsiaCell is relatively higher by 0.008. However, a result can be statistically significant but still may be too small to be practically useful. The negative signs of the coefficients in the education categories imply that educated people in the region are likely to prefer AsiaCell service over other services.

The probability of using Korek rather than AsiaCell among students and those who work in the private and public sectors relative to those who have no permanent job is higher by 0.473, 0.431, and 0.585, respectively. Korek’s high-quality network makes the service desirable. In addition, because of the poor infrastructure of and the continuous development program of the region, public employees are provided with credits by the government to use mobile telecommunication in performing their jobs.

The probability of using Korek rather than AsiaCell for social purposes is higher by 0.070. Korek is trying to retain customers by offering more entertainment features. The probability of using the other service providers rather than AsiaCell for entertainment is higher by 0.023. The other service providers are trying to attract more customers by offering different entertainment features.

The second model in this study analyzes the factors that are relevant to the purpose of using mobile telecommunication. The negative coefficient among men indicates that the marginal effect of this variable to the probability of using the service mainly for social purposes is negative. In other words, women are more likely to use the service for social networking.

The negative coefficients of age imply that the probability of using the service for social and entertainment purposes are lower among younger subscribers. This is likely due to the high cost of International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2305

telecommunication services in the region and the typically low income of younger people. Younger people use the service mainly for work-related purposes.

Positive coefficients of media, battery, and color for entertainment use are relative to work purposes. This indicates that subscribers who prefer to have a cell phone with better media quality, long battery life, and more color options are more likely to use the service for entertainment than for work- related tasks. Moreover, the positive coefficient of Korean brands and other brands relative to Nokia for entertainment use is an indication that Korean cell phones contain more entertainment features than Nokia brands.

Summary, Conclusion, and Policy Implications

This study contributes to the literature in several ways. First, it finds evidence of leapfrogging and diffusion of ICT systems and services in a landlocked region under rapid development. Second, it sheds on the heterogeneity in the ICT sector development, planned and managed regionally by a late-late comer vs. centrally. Third, it shows the power of new ICT technology and how it crowds out the old technologies. Fourth, it shows the evolution of new Kurdish regional and national corporations and their entry into the competitive Middle Eastern ICT market as main players.

The two multinomial logit models provided evidence of the overall significance of the explanatory variables. The models were found to have a high explanatory power, and individual parameters were strongly significant with the expected signs. This indicates that the model as a whole explains quite well customers’ intentions to use mobile services, and the analyses and tests show evidence of reliability of the factors measuring information technology adoption. Despite a lack of previous empirical studies in Kurdistan, we have developed a well-established scale for measuring the intentions to adopt information technology that can be used by business managers, policy makers, and regulators.

These findings prove the relevance of the proposed and successfully tested hypotheses. The study represents a critical step toward developing a technology acceptance theory in Kurdistan’s mobile telecommunication sector. The study also presents important theoretical and practical contributions useful to market analyzers and decision makers. It provides a useful conceptualization of the formation of technology acceptance through examining its determinants.

The infrastructure in Kurdistan has been neglected for years as a result of sanctions and internal strife. In the short term, the enablement of ICT in the region is as much of a development issue as it is a policy issue. Initial policy development should form the basis of a policy environment that will facilitate rapid growth and exploitation of the telecommunication industry. To catch up in the long run, the regional government must create a suitable regulatory environment, act to create international connectivity with neighboring states, and form partnerships with businesses to ensure an optimized network and with the education sector to develop necessary ICT skills among workers.

Mobile phones are already used extensively within the region. However, the quality of the services is not monitored by the government, and a proper regulatory policy needs to be developed. Regulation of the market in the region will encourage firms to build and maintain a competitive advantage, achieve higher customer satisfaction, and improve service quality. This study is a first attempt to assess— 2306 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

and suggests an initiative to regularly reassess—service quality to meet customer expectations and provide satisfaction.

The estimated models show that Korek is the preferred service among the available services. To attract more customers, AsiaCell should address the weak points of its service such as network quality and available features. However, according to our results, the younger generation prefers to use Korek over AsiaCell and other providers; therefore, Korek should pay more attention to this age cohort, extending service offers and identifying the factors affecting the nonuse of the service by older age cohorts.

Future research should be forward looking and focused on the examination of similar models with a specific service such as SMS, MMS, Internet service, customized service, and customer service. Additional research is also needed to define better measures of calling cost and cell phone prices.

International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2307

References

Aboelmaged, M. G. (2010). Predicting e-procurement adoption in a developing country: An empirical integration of technology acceptance model and theory of planned behaviour. Industrial Management & Data Systems, 110(3–4), 392–414. doi:10.1108/02635571011030042

Adams, D. A., Nelson, R. R., & Todd, P. A. (1992). Perceived usefulness, ease of use, and usage of information technology—A replication. MIS Quarterly, 16(2), 227–247. doi:10.2307/249577

Agarwal, R., & Prasad, J. (1997). The role of innovation characteristics and perceived voluntariness in the acceptance of information technologies. Decision , 28(3), 557–582. doi:10.1111/j.1540- 5915.1997.tb01322.x

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. doi:10.1016/0749-5978(91)90020-t

Alborz, B. (2009). Value-base adoption of mobile Internet in (MSc thesis), Lulea University of Technology, Lulea, Sweden.

AlHinai, Y. S., Kurnia, S., & Johnston, R. B. (2007, July). Adoption of mobile commerce services by individuals: A meta-analysis of the litrature. Paper presented at the Sixth International Conference on the Management of Mobile Business (ICMB 2007). Toronto, ON, Canada.

Al-Qirim, N. (2007). The adoption and diffusion of e-commerce in developing countries: The case of an NGO in Jordan. Information Technology for Development, 13(2), 107–131. doi:10.1002/itdj.20053

Al-Shafi, S., & Weerakkody, V. (2009, [April]). Factors affecting e-government adoption in the state of Qatar. Paper presented at the European and Mediterranean Conference on Information Systems, Abu Dhabi, .

Avgerou C. (2009). Discourses on innovation and development in information systems in developing countries research. In E. Byrne, B. Nicholson, & E. Salem (Eds.), Assessing the Contribution of ICT to Development Goals, Dubai School of Government: UAE; 1–21. Retrieved from http://www.ifip.dsg.ae

Bagozzi, R. P. (2008). The legacy of the technology acceptance model and a proposal for a paradigm shift. Journal of the Association for Information Systems, 8(4), 244–254.

Baker, P. M. A., & Bellordre, C. (2003). Factors influencing adoption of technologies: Key issues, barriers and opportunities for people with disabilities. [Los Angeles, CA]: Rehabilitation Research Center for Wireless Technologies. 2308 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

Bhattacherjee, A., & Premkumar, G. (2004). Understanding changes in belief and attitude toward information technology usage: A theoretical model and longitudinal test. MIS Quarterly, 28(2), 229–254. doi:10.2307/25148634

Bouwman, H., & van de Wijngaert, L. (2009). context, and conjoints: A reassessment of TAM. Journal of Information Technology, 24(2), 186–201. doi:10.1057/Jit.2008.36

Burton-Jones, A., & Hubona, G. S. (2006). The mediation of external variables in the technology acceptance model. Information & Management, 43(6), 706–717. doi:10.1016/J,Im.2006.03.007

Bwalya, K. J. (2009). Factors affecting adoption of e-government in zambia. Electronic Journal of Information Systems in Developing Countries, 38(4), 1–13.

Carlsson, C., Hyvonen, K., Repo, P., & Walden, P. (2005, June). Adoption of mobile services across different platforms. Paper presented at the the 18th Bled eCommerce Conference, Bled, Slovenia.

Chuttur, M. Y. (2009). Overview of the technology acceptance model: Origins, developments and future directions. Sprouts: Working Papers on Information Systems 9(37), 1–23. Retrieved from http://sprouts.aisnet.org/9-37

Coursaris, C., Hassanein, K., Head, M., & Bontis, N. (2007, June). The impact of distractions on the usability and the adoption of mobile devices for wireless data services. Paper presented at the the 15th European Conference on Information Systems (ECIS 2007), St. Gallen, Switzerland.

Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. doi:10.2307/249008

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of technology: A comparison of two theoretical models. Management , 35(8), 982–1003. doi:10.2307/2632151

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1992). Extrinsic and intrinsic motivation to use in the workplace. Journal of Applied Social Psychology, 22(14), 1111–1132. doi:10.1111/j.1559- 1816.1992.tb00945.x

Dercon, S., & Christiaensen, L. (2011). Consumption risk, technology adoption and poverty traps: Evidence from Ethiopia. Journal of Development Economics, 96(2), 159–173. doi:10.1016/j.jdeveco.2010.08.003

Duncombe, R., & Boateng, R. (2009). Mobile phones and financial services in developing countries: A review of concepts, methods, issues, evidence and future research directions. Third World Quarterly, 30(7), 1237–1258. doi:10.1080/01436590903134882 International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2309

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. , MA: Addison-Wesley.

Fong, M. W. L. (2009). : The case of developing countries. Issues in Informing Science and Information Technology, 6(2), 471–478.

Foster, A. D., & Rosenzweig, M. R. (2010). of technology adoption. Annual Review of Economics, 2(1), 395–424. doi:10.1146/annurev.economics.102308.124433

Giné, X., & Yang, D. (2009). Insurance, credit, and technology adoption: Field experimental evidencefrom Malawi. Journal of Development Economics, 89(1), 1–11. doi:10.1016/j.jdeveco.2008.09.007

Greene, W. (2008). Econometric analysis (6th ed.), Upper Saddle River, NJ: Prentice Hall.

Gujarati, D. N. (2004). Basic econometrics (4th ed.). Cambridge, UK: Cambridge University Press.

Gulbahar, Y., & Guven, I. (2008). A survey on ICT usage and the perceptions of social studies teachers in Turkey. Educational Technology & Society, 11(3), 37–51.

Hong, S. J., & Tam, K. Y. (2006). Understanding the adoption of multipurpose information appliances: The case of mobile data services. Information Systems Research, 17(2), 162–179. doi:10.1287/isre.1060.0088

Howard, P. N., & Mazaheri, N. (2009). reform, Internet use and mobile phone adoption in the developing world. World Development, 37(7), 1159–1169. doi:10.1016/j.worlddev.2008.12.005

Hsu, C. L., Lu, H. P., & Hsu, H. H. (2007). Adoption of the mobile Internet: An empirical study of multimedia message service (MMS). Omega-International Journal of Management Science, 35(6), 715–726. doi:10.1016/j.omega.2006.03.005

International Telecommunication Union. (2012). Connect Arab Summit 2012: Statistical factsheet. Geneva, Switzerland: Author. Retrieved from http://www.itu.int/net/newsroom/connect/arab/2012/reports/statistical_overview.aspx

Khayyat, N. T. (2010). Effects of information technology on cost, quality and efficiency in provision of public services. In A. Heshmati & S. Peng (Eds.), Information and communication technologies policies and practices (pp.73–90). New York: NOVA Science Publishers.

Khayyat, N. T., & Heshmati, A. (2012). Determinant of mobile phone customer satisfaction in Kurdistan region. Journal of Knowledge Management, Economics and Information Policy, 2(3), 91–121.

2310 Nabaz T. Khayyat & Almas Heshmati International Journal of Communication 7(2013)

Kijsanayotin, B., Pannarunothai, S., & Speedie, S. M. (2009). Factors influencing health information technology adoption in Thailand’s community health centers: Applying the UTAUT model. Internatinal Journal of Medical Informatics, 78(6), 404–416. doi:10.1016/j.ijmedinf.2008.12.005

Kim, H. W., Chan, H. C., & Gupta, S. (2007). Value-based adoption of mobile Internet: An empirical investigation. Decision Support Systems, 43(1), 111–126. doi:10.1016/j.dss.2005.05.009

Kim, S. S., & Malhotra, N. K. (2005). A longitudinal model of continued IS use: An integrative view of four mechanisms underlying postadoption phenomena. Management Science, 51(5), 741–755. doi:10.1287/mnsc.1040.0326

King, W. R., & He, J. (2006). A meta-analysis of the technology acceptance model. Information & Management, 43(6), 740–755. doi:10.1016/J.Im.2006.05.003

Lee, I., Choi, B., Kim, J., & Hong, S. J. (2007). Culture-technology fit: Effects of cultural characteristics on the post-adoption beliefs of mobile Internet users. International Journal of Electronic Commerce, 11(4), 11–51. doi:10.2753/Jec1086-4415110401

Lee, W. J., Kim, T. U., & Chung, J-Y. (2002, July). User acceptance of the mobile Internet in m-business. Paper presented at the the First International Conference on Mobile Business. Athens, Greece.

Liao, C. H., Tsou, C. W., & Huang, M. F. (2007). Factors influencing the usage of mobile services in Taiwan. Online Information Review, 31(6), 759–774. doi:10.1108/14684520710841757

Lu, J., Yao, J. E., & Yu, C. S. (2005). Personal innovativeness, social influences and adoption of wireless Internet services via mobile technology. Journal of Strategic Information Systems, 14(3), 245– 268. doi:10.1016/j.jsis.2005.07.003

Marcopolis. (2012). Telecommunication companies in Iraq and Kurdistan: Telecom list. Retrieved from http://www.marcopolis.net/telecommunication-companies-in-iraq-kurdistan-telecom-list.htm

Moon, J. W., & Kim, Y. G. (2001). Extending the TAM for a World-Wide-Web context. Information & Management, 38(4), 217–230. doi:10.1016/S0378-7206(00)00061-6

Moore, G. C., & Benbasat, I. (1991). Development of an instrument to measure the perceptions of adopting an information technology innovation. Information Systems Research, 2(3), 192–222.

Ndubisi, N. O., Gupta, O. K., & Massoud, S. (2003). Organizational learning and vendor support quality by the usage of application software packages: A study of Asian entrepreneurs. Journal of Systems Science and Systems Engineering, 12(3), 314–331. doi:10.1007/s11518-006-0138-2

Pagani, M. (2004). Determinants of adoption of third generation mobile multimedia services. Journal of Interactive Marketing, 18(3), 46–59. doi:10.1002/dir.20011 International Journal of Communication 7 (2013) Mobile Telecommunication in Kurdistan 2311

Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1985). A conceptual model of service quality and its implications for future research. Journal of Marketing, 49(4), 41–50. doi:10.2307/1251430

Park, N., Roman, R., Lee, S., & Chung, J. E. (2009). User acceptance of a digital library system in developing countries: An application of the technology acceptance model. International Journal of Information Management, 29(3), 196–209. doi:10.1016/j.ijinfomgt.2008.07.001

Rogers, E. M. (1995). Diffusion of innovations. New York, NY: Free Press.

Roos, I., & Gustafsson, A. (2007). Understanding frequent switching patterns: A crucial element in managing customer relationships. Journal of Service Research, 10(1), 93–108. doi:10.1177/1094670507303232

Sanford, C., & Oh, H. (2010). The role of user resistance in the adoption of a mobile data service. Cyberpsychology, Behavior, and Society Networking, 13(6), 663–672. doi:10.1089/cyber.2009.0377

Train, K. (2009). Discrete choice methods with simulation (2nd ed.). New York, NY: Cambridge University Press.

United Nations Development Programme. (2004). Iraq living conditions survey tabulation report (Vol. 1). New York, NY: Author. van der Heijden, H. (2003). Factors influencing the usage of : The case of a generic portal in The . Information & Management, 40(6), 541–549. doi:10.1016/s0378-7206(02)00079-4

Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science, 46(2), 186–204. doi:10.1287/mnsc.46.2.186.11926

Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425–478. doi:10.2307/30036540

Wei, T. T., Marthandan, G., Chong, A. Y-L., Ooi, K-B., & Arumugam, S. (2009). What drives Malaysian m- commerce adoption? An empirical analysis. Industrial Management & Data Systems, 109(3), 370–388.

Yang, K., & Jolly, L. D. (2009). The effects of consumer perceived value and subjective norm on mobile data service adoption between American and Korean consumers. Journal of Retailing and Consumer Services, 16(6), 502–508. doi:10.1016/j.jretconser.2009.08.00